Adapt to Elster

Change response template to answer in German.
Change storage location to sroecker/Elster-preference.

app.py

@@ -59,7 +59,7 @@ else:
     print(f"Creating new dataset file: {dataset_file}")
 
 # Set up CommitScheduler for dataset uploads
-repo_id = "davanstrien/magpie-preference"  # Replace with your desired dataset repo
+repo_id = "sroecker/Elster-preference"  # Replace with your desired dataset repo
 scheduler = CommitScheduler(
     repo_id=repo_id,
     repo_type="dataset",
@@ -151,7 +151,7 @@ def generate_instruction_response():
         gr.update(interactive=False),
     )
 
-    response_template = f"""<|begin_of_text|><|start_header_id|>user<|end_header_id|>\n\n{sanitized_instruction}<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"""
+    response_template = f"""<|begin_of_text|><|start_header_id|>user<|end_header_id|>\n\n{sanitized_instruction} Antworte auf Deutsch ohne "Sie".<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"""
 
     response = pipeline(
         response_template,
@@ -190,11 +190,11 @@ def generate_instruction_response():
 
 
 title = """
-<h1 style="text-align:center">&#x1F426; Magpie Preference</h1>
+<h1 style="text-align:center">&#x1F426; Elster Preference</h1>
 """
 
 description = """
-This demo showcases **[Magpie](https://magpie-align.github.io/)**, an innovative approach to generating high-quality data by prompting aligned LLMs with their pre-query templates. Unlike many existing synthetic data generation methods, Magpie doesn't rely on prompt engineering or seed questions for generating synthetic data. Instead, it uses the prompt template of an aligned LLM to generate both the user query and an LLM response.
+This demo showcases **Elster** - derived from **[Magpie](https://magpie-align.github.io/)**, an innovative approach to generating high-quality data by prompting aligned LLMs with their pre-query templates. Unlike many existing synthetic data generation methods, Magpie doesn't rely on prompt engineering or seed questions for generating synthetic data. Instead, it uses the prompt template of an aligned LLM to generate both the user query and an LLM response.
 
 <img src="https://magpie-align.github.io/images/pipeline.png" alt="Magpie Pipeline" width="50%" align="center" />